def backtest_PERandPBR(mainParament):
Temp_reset = 0 #休息日剩餘天數
Temp_result_pick = pd.DataFrame(columns=['date', '選股數量'])
userInfo = get_user_info.data_user_info(mainParament.money_start,
mainParament.date_start,
mainParament.date_end)
while userInfo.now_day <= userInfo.end_day:
#週末直接跳過
if userInfo.now_day.isoweekday() in [6, 7]:
print(
str(userInfo.now_day) + 'is 星期' +
str(userInfo.now_day.isoweekday()))
if userInfo.add_one_day() == False: #加一天
break
continue
#先看看台積有沒有資料，如果沒有表示這天是非週末假日跳過
if get_stock_history.get_stock_price(
2330, userInfo.now_day,
get_stock_history.stock_data_kind.AdjClose) == None:
print(str(userInfo.now_day) + "這天沒開市")
if userInfo.add_one_day() == False: #加一天
break
continue
#休息日直接跳過
if Temp_reset > 0 and len(userInfo.handle_stock) == 0:
print(
str(userInfo.now_day) + ' is reset time:第' + str(Temp_reset) +
'天')
if userInfo.add_one_day() == False: #加一天
break
Temp_reset = Temp_reset - 1
continue
#開始篩選--------------------------------------
Temp_result0 = {}
Temp_result = pd.DataFrame()
if bool_check_PER_pick: #PER pick
Temp_result0['PER'] = get_stock_history.get_PER_range(
userInfo.now_day, mainParament.PER_start, mainParament.PER_end)
if bool_check_PBR_pick: #PBR pick
Temp_result0['PBR'] = get_stock_history.get_PBR_range(
userInfo.now_day, mainParament.PBR_start, mainParament.PBR_end)
Temp_result = tools.MixDataFrames(Temp_result0)
#出場訊號篩選--------------------------------------
if len(Temp_result) < mainParament.Pick_amount and len(
userInfo.handle_stock) > 0:
userInfo.sell_all_stock()
Temp_reset = mainParament.change_days
#入場訊號篩選--------------------------------------
if len(Temp_result
) > mainParament.Pick_amount and Temp_reset == 0 and len(
userInfo.handle_stock) == 0:
Temp_buy0 = {'result': Temp_result}
if bool_check_price_pick:
Temp_buy0['price'] = get_stock_history.get_price_range(
userInfo.now_day, mainParament.price_high,
mainParament.price_low, Temp_result)
Temp_buy0['price'] = Temp_buy0['price'].sort_values(
by='price', ascending=False)
if bool_check_volume_pick:
Temp_buy0['volume'] = get_stock_history.get_AVG_value(
userInfo.now_day, mainParament.volumeAVG,
mainParament.volumeDays, Temp_result)
Temp_buy0['volume'] = Temp_buy0['volume'].sort_values(
by='volume', ascending=False)
Temp_buy = tools.MixDataFrames(Temp_buy0)
if Temp_buy0['price'].empty == False:
Temp_buy = Temp_buy.sort_values(by='price', ascending=False)
if Temp_buy0['volume'].empty == False:
Temp_buy = Temp_buy.sort_values(by='volume', ascending=False)
userInfo.buy_all_stock(Temp_buy)
Temp_reset = mainParament.change_days
#更新資訊--------------------------------------
userInfo.Record_userInfo()
userInfo.Recod_tradeInfo()
Temp_result_pick = Temp_result_pick.append(
{
'date': userInfo.now_day,
'選股數量': len(Temp_result)
},
ignore_index=True)
#加一天----------------------------
if userInfo.add_one_day() == False:
break
#最後總結算----------------------------
Temp_result_pick.set_index('date', inplace=True)
Temp_alldata = tools.MixDataFrames(
{
'draw': userInfo.Temp_result_draw,
'pick': Temp_result_pick
}, 'date')
draw_figur.draw_backtest(userInfo.Temp_result_draw)
Temp_alldata = tools.MixDataFrames(
{
'all': Temp_alldata,
'userinfo': userInfo.Temp_result_All
}, 'date')
userInfo.Temp_result_draw.to_csv('backtestdata.csv')
userInfo.Temp_trade_info.to_csv('backtesttrade.csv')
Temp_alldata.to_csv('backtestAll.csv')
def backtest_KD_pick(mainParament):
userInfo = get_user_info.data_user_info(mainParament.money_start,
mainParament.date_start,
mainParament.date_end)
Temp_result_pick = pd.DataFrame(columns=['date', '選股數量'])
Temp_table = get_stock_history.get_stock_history(2330,
userInfo.start_day,
reGetInfo=False,
UpdateInfo=False)
All_stock_signal = dict()
for key, value in get_stock_info.ts.codes.items():
if value.market == "上市" and len(value.code) == 4:
if get_stock_history.check_no_use_stock(value.code) == True:
print('get_stock_price: ' + str(value.code) + ' in no use')
continue
table = get_stock_history.get_stock_history(value.code,
userInfo.start_day,
reGetInfo=False,
UpdateInfo=False)
table_K, table_D = talib.STOCH(table['High'],
table['Low'],
table['Close'],
fastk_period=50,
slowk_period=20,
slowk_matype=0,
slowd_period=20,
slowd_matype=0)
table_sma10 = talib.SMA(np.array(table['Close']), 10)
table_sma240 = talib.SMA(np.array(table['Close']), 240)
signal_buy = (table_K > table_D)
signal_sell = (table_K < table_D)
signal_sma10 = table.Close < table_sma10
signal_sma240 = table.Close > table_sma240
signal = signal_buy.copy()
signal = (signal_sma10 & signal_sma240 & signal_buy)
signal[signal_sell] = -1
All_stock_signal[value.code] = signal
ROE_record_day = datetime.datetime(2000, 1, 1)
buy_data = pd.DataFrame(columns=['Date', '公司代號']).set_index('Date')
sell_data = pd.DataFrame(columns=['Date', '公司代號']).set_index('Date')
ROE_data = {}
jump_day = 240
for index, row in Temp_table.iterrows():
while userInfo.now_day < index:
userInfo.add_one_day()
if jump_day > 0:
jump_day = jump_day - 1
continue
if ROE_record_day <= index:
ROE_data = {}
ROE_record_day = tools.changeDateMonth(index, 1)
BOOK_data = get_stock_history.get_allstock_financial_statement(
tools.changeDateMonth(index, -3), get_stock_history.FS_type.BS)
CPL_data = get_stock_history.get_allstock_financial_statement(
tools.changeDateMonth(index, -3),
get_stock_history.FS_type.CPL)
ROE_data["ROE_data_1"] = pd.DataFrame({
'ROE_data_1':
(CPL_data["本期綜合損益總額（稅後）"] / BOOK_data['權益總額']) * 100
})
BOOK_data = get_stock_history.get_allstock_financial_statement(
tools.changeDateMonth(index, -6), get_stock_history.FS_type.BS)
CPL_data = get_stock_history.get_allstock_financial_statement(
tools.changeDateMonth(index, -6),
get_stock_history.FS_type.CPL)
ROE_data["ROE_data_2"] = pd.DataFrame({
'ROE_data_2':
(CPL_data["本期綜合損益總額（稅後）"] / BOOK_data['權益總額']) * 100
})
BOOK_data = get_stock_history.get_allstock_financial_statement(
tools.changeDateMonth(index, -9), get_stock_history.FS_type.BS)
CPL_data = get_stock_history.get_allstock_financial_statement(
tools.changeDateMonth(index, -9),
get_stock_history.FS_type.CPL)
ROE_data["ROE_data_3"] = pd.DataFrame({
'ROE_data_3':
(CPL_data["本期綜合損益總額（稅後）"] / BOOK_data['權益總額']) * 100
})
BOOK_data = get_stock_history.get_allstock_financial_statement(
tools.changeDateMonth(index, -12),
get_stock_history.FS_type.BS)
CPL_data = get_stock_history.get_allstock_financial_statement(
tools.changeDateMonth(index, -12),
get_stock_history.FS_type.CPL)
ROE_data["ROE_data_4"] = pd.DataFrame({
'ROE_data_4':
(CPL_data["本期綜合損益總額（稅後）"] / BOOK_data['權益總額']) * 100
})
mask = tools.MixDataFrames(ROE_data)
ROE_data_result = (mask["ROE_data_1"] + mask["ROE_data_2"] +
mask["ROE_data_3"] + mask["ROE_data_4"]) / 4
ROE_data_result = ROE_data_result.dropna()
ROE_data = mask["ROE_data_1"] > ROE_data_result
buy_numbers = []
sell_numbers = []
for i, value in All_stock_signal.items():
try:
if value[index] == True:
if ROE_data.index.__contains__(int(i)):
if ROE_data[int(i)] == True:
buy_numbers.append(i)
elif value[index] < 0:
sell_numbers.append(i)
except:
#print('error:' + str(index) + ' at ' + str(i))
continue
#出場訊號篩選--------------------------------------
if len(userInfo.handle_stock) > 0:
Temp_data = userInfo.handle_stock
for key, value in list(Temp_data.items()):
if sell_numbers.__contains__(key):
userInfo.sell_stock(key, value.amount)
#入場訊號篩選--------------------------------------
if len(buy_numbers) > 0:
Temp_buy = pd.DataFrame(columns={'公司代號', 'volume'})
for number in buy_numbers:
volume = get_stock_history.get_stock_price(
number, tools.DateTime2String(userInfo.now_day),
get_stock_history.stock_data_kind.Volume)
Temp_buy = Temp_buy.append({
'公司代號': number,
'volume': volume
},
ignore_index=True)
Temp_buy = Temp_buy.sort_values(by='volume',
ascending=False).set_index('公司代號')
userInfo.buy_all_stock(Temp_buy)
if len(buy_numbers) != 0:
buy_data = buy_data.append({
'Date': index,
'公司代號': buy_numbers
},
ignore_index=True)
if len(sell_numbers) != 0:
sell_data = sell_data.append({
'Date': index,
'公司代號': sell_numbers
},
ignore_index=True)
#更新資訊--------------------------------------
userInfo.Record_userInfo()
userInfo.Recod_tradeInfo()
Temp_result_pick = Temp_result_pick.append(
{
'date': userInfo.now_day,
'選股數量': len(buy_numbers)
},
ignore_index=True)
buy_data = buy_data.set_index('Date')
buy_data.to_csv('buy.csv')
sell_data = sell_data.set_index('Date')
sell_data.to_csv('sell.csv')
#最後總結算----------------------------
Temp_result_pick.set_index('date', inplace=True)
Temp_alldata = tools.MixDataFrames(
{
'draw': userInfo.Temp_result_draw,
'pick': Temp_result_pick
}, 'date')
draw_figur.draw_backtest(userInfo.Temp_result_draw.set_index('date'))
Temp_alldata = tools.MixDataFrames(
{
'all': Temp_alldata,
'userinfo': userInfo.Temp_result_All
}, 'date')
userInfo.Temp_result_draw.set_index('date').to_csv('backtestdata.csv')
userInfo.Temp_trade_info.set_index('date').to_csv('backtesttrade.csv')
Temp_alldata.set_index('date').to_csv('backtestAll.csv')
def backtest_Regular_quota(mainParament):
userInfo = get_user_info.data_user_info(0, mainParament.date_start,
mainParament.date_end)
buy_month = mainParament.date_start
Temp_result_pick = pd.DataFrame(columns=['date', '選股數量'])
while userInfo.now_day <= userInfo.end_day:
#週末直接跳過
if userInfo.now_day.isoweekday() in [6, 7]:
print(
str(userInfo.now_day) + 'is 星期' +
str(userInfo.now_day.isoweekday()))
if userInfo.add_one_day() == False: #加一天
break
continue
#先看看台積有沒有資料，如果沒有表示這天是非週末假日跳過
if get_stock_history.get_stock_price(
2330, userInfo.now_day,
get_stock_history.stock_data_kind.AdjClose) == None:
print(str(userInfo.now_day) + "這天沒開市")
if userInfo.add_one_day() == False: #加一天
break
continue
#開始篩選--------------------------------------
Temp_result = pd.DataFrame()
Temp_result = Temp_result.append(
{
'date': userInfo.now_day,
'number': mainParament.buy_number
},
ignore_index=True)
#入場訊號篩選--------------------------------------
if userInfo.now_day.month != buy_month.month and userInfo.now_day.day >= mainParament.buy_day:
Temp_price = get_stock_history.get_stock_price(
mainParament.buy_number, userInfo.now_day,
get_stock_history.stock_data_kind.AdjClose)
userInfo.now_money = userInfo.now_money + mainParament.money_start
userInfo.start_money = userInfo.start_money + mainParament.money_start
Temp_stockNumber = tools.Count_Stock_Amount(
mainParament.money_start, Temp_price)
userInfo.buy_stock(mainParament.buy_number, Temp_stockNumber)
buy_month = userInfo.now_day
#更新資訊--------------------------------------
if userInfo.start_money > 0:
userInfo.Record_userInfo()
userInfo.Recod_tradeInfo()
Temp_result_pick = Temp_result_pick.append(
{
'date': userInfo.now_day,
'選股數量': len(Temp_result)
},
ignore_index=True)
#加一天----------------------------
if userInfo.add_one_day() == False:
break
#最後總結算----------------------------
Temp_result_pick.set_index('date', inplace=True)
Temp_alldata = tools.MixDataFrames(
{
'draw': userInfo.Temp_result_draw,
'pick': Temp_result_pick
}, 'date')
draw_figur.draw_backtest(userInfo.Temp_result_draw)
Temp_alldata = tools.MixDataFrames(
{
'all': Temp_alldata,
'userinfo': userInfo.Temp_result_All
}, 'date')
userInfo.Temp_result_draw.to_csv('backtestdata.csv')
userInfo.Temp_trade_info.to_csv('backtesttrade.csv')
Temp_alldata.to_csv('backtestAll.csv')
def backtest_Record_high(mainParament):
Temp_reset = 0 #休息日剩餘天數
userInfo = get_user_info.data_user_info(mainParament.money_start,
mainParament.date_start,
mainParament.date_end)
Temp_result_pick = pd.DataFrame(columns=['date', '選股數量'])
while userInfo.now_day <= userInfo.end_day:
#週末直接跳過
if userInfo.now_day.isoweekday() in [6, 7]:
print(
str(userInfo.now_day) + 'is 星期' +
str(userInfo.now_day.isoweekday()))
if userInfo.add_one_day() == False: #加一天
break
continue
#先看看台積有沒有資料，如果沒有表示這天是非週末假日跳過
if get_stock_history.get_stock_price(
2330, userInfo.now_day,
get_stock_history.stock_data_kind.AdjClose) == None:
print(str(userInfo.now_day) + "這天沒開市")
if userInfo.add_one_day() == False: #加一天
break
continue
#休息日直接跳過
if Temp_reset > 0 and len(userInfo.handle_stock) == 0:
print(
str(userInfo.now_day) + ' is reset time:第' + str(Temp_reset) +
'天')
if userInfo.add_one_day() == False: #加一天
break
Temp_reset = Temp_reset - 1
continue
#出場訊號篩選--------------------------------------
if len(userInfo.handle_stock) > 0:
Temp_data = userInfo.handle_stock
for key, value in list(Temp_data.items()):
if get_stock_history.get_stock_price(
key, userInfo.now_day,
get_stock_history.stock_data_kind.AdjClose
) < get_stock_history.get_stock_MA(key, userInfo.now_day, 20):
userInfo.sell_stock(key, value.amount)
#開始篩選--------------------------------------
Temp_result = pd.DataFrame()
Temp_result0 = {}
if Temp_reset <= 0:
if bool_check_ROE_pick == True:
Temp_result0['ROE'] = get_stock_history.get_ROE_range(
userInfo.now_day, mainParament.ROE_start,
mainParament.ROE_end)
if bool_check_ROE_pick == True:
Temp_result0[
'ROE_last_seson'] = get_stock_history.get_ROE_range(
userInfo.now_day - timedelta(weeks=4), 1, 10000)
Temp_result = tools.MixDataFrames(Temp_result0)
#入場訊號篩選--------------------------------------
if Temp_reset <= 0 and len(Temp_result) > 0:
Temp_buy0 = {'result': Temp_result}
if bool_check_PBR_pick:
Temp_buy0['PBR'] = get_stock_history.get_PBR_range(
userInfo.now_day, mainParament.PBR_start,
mainParament.PBR_end)
if bool_check_price_pick:
Temp_buy0['price'] = get_stock_history.get_price_range(
userInfo.now_day, mainParament.price_high,
mainParament.price_low, Temp_result)
if bool_check_volume_pick:
Temp_buy0['volume'] = get_stock_history.get_AVG_value(
userInfo.now_day, mainParament.volumeAVG,
mainParament.volumeDays, Temp_result)
Temp_buy0['point'] = pd.DataFrame(columns=['公司代號', 'point'])
for index, row in Temp_result.iterrows():
try:
temp = Temp_buy0['PBR'].at[index, 'PBR']
except:
continue
if get_stock_history.check_no_use_stock(index):
continue
Temp_point = (row['ROE_x'] /
row['ROE_y']) / Temp_buy0['PBR'].at[index, 'PBR']
Temp_buy0['point'] = Temp_buy0['point'].append(
{
'公司代號': index,
'point': Temp_point
}, ignore_index=True)
Temp_buy0['point'] = Temp_buy0['point'].astype({
'公司代號': 'int',
'point': 'float'
})
Temp_buy0['point'] = Temp_buy0['point'].set_index('公司代號')
Temp_buy = tools.MixDataFrames(Temp_buy0)
Temp_buy1 = {'result': Temp_buy}
Temp_buy1['high'] = get_stock_history.get_RecordHigh_range(
userInfo.now_day, mainParament.change_days,
mainParament.Record_high_day, Temp_buy)
Temp_buy = tools.MixDataFrames(Temp_buy1)
if Temp_buy.empty == False:
Temp_buy = Temp_buy.sort_values(by='point', ascending=False)
Temp_buy = Temp_buy.head(3)
userInfo.buy_all_stock(Temp_buy)
#更新資訊--------------------------------------
if Temp_reset <= 0:
Temp_reset = mainParament.change_days
userInfo.Record_userInfo()
userInfo.Recod_tradeInfo()
Temp_result_pick = Temp_result_pick.append(
{
'date': userInfo.now_day,
'選股數量': len(Temp_result)
},
ignore_index=True)
#加一天----------------------------
if userInfo.add_one_day() == False:
break
else:
Temp_reset = Temp_reset - 1
#最後總結算----------------------------
Temp_result_pick.set_index('date', inplace=True)
Temp_alldata = tools.MixDataFrames(
{
'draw': userInfo.Temp_result_draw,
'pick': Temp_result_pick
}, 'date')
draw_figur.draw_backtest(userInfo.Temp_result_draw)
Temp_alldata = tools.MixDataFrames(
{
'all': Temp_alldata,
'userinfo': userInfo.Temp_result_All
}, 'date')
userInfo.Temp_result_draw.to_csv('backtestdata.csv')
userInfo.Temp_trade_info.to_csv('backtesttrade.csv')
Temp_alldata.to_csv('backtestAll.csv')
def backtest_monthRP_Up(mainParament):
Temp_change = 0#換股剩餘天數
userInfo = get_user_info.data_user_info(mainParament.money_start,mainParament.date_start,mainParament.date_end)
Temp_result_pick = pd.DataFrame(columns=['date','選股數量'])
while userInfo.now_day <= userInfo.end_day:
#週末直接跳過
if userInfo.now_day.isoweekday() in [6,7]:
print(str(userInfo.now_day) + 'is 星期' + str(userInfo.now_day.isoweekday()))
if userInfo.add_one_day() == False:#加一天
break
continue
#先看看台積有沒有資料，如果沒有表示這天是非週末假日跳過
if get_stock_history.get_stock_price(2330,userInfo.now_day,get_stock_history.stock_data_kind.AdjClose) == None:
print(str(userInfo.now_day) + "這天沒開市")
if userInfo.add_one_day() == False:#加一天
break
continue
#出場訊號篩選--------------------------------------
if Temp_change <= 0 and len(userInfo.handle_stock) > 0:
userInfo.sell_all_stock()
#開始篩選--------------------------------------
Temp_result0 = {}
Temp_result = pd.DataFrame()
if Temp_change <= 0:
if bool_check_monthRP_pick:#月營收升高篩選(月為單位)
Temp_result0['month'] = get_stock_history.get_monthRP_up(userInfo.now_day,mainParament.smoothAVG,mainParament.upMonth)
if bool_check_ROE_pick:#ROE
Temp_result0['ROE'] = get_stock_history.get_ROE_range(userInfo.now_day,mainParament.ROE_start,mainParament.ROE_end)
if bool_check_PBR_pick:#PBR
Temp_result0['PBR'] = get_stock_history.get_PBR_range(userInfo.now_day,mainParament.PBR_start,mainParament.PBR_end)
if bool_check_PER_pick:#PER
Temp_result0['PER'] = get_stock_history.get_PER_range(userInfo.now_day,mainParament.PER_start,mainParament.PER_end)
Temp_result = tools.MixDataFrames(Temp_result0)
#入場訊號篩選--------------------------------------
if Temp_change <= 0 and len(userInfo.handle_stock) <= 0 and len(Temp_result) > mainParament.Pick_amount:
Temp_buy0 = {'result':Temp_result}
if bool_check_price_pick:
Temp_buy0['price'] = get_stock_history.get_price_range(userInfo.now_day,mainParament.price_high,mainParament.price_low,Temp_result)
Temp_buy0['price'] = Temp_buy0['price'].sort_values(by='price', ascending=False)
if bool_check_volume_pick:
Temp_buy0['volume'] = get_stock_history.get_AVG_value(userInfo.now_day,mainParament.volumeAVG,mainParament.volumeDays,Temp_result)
Temp_buy0['volume'] = Temp_buy0['volume'].sort_values(by='volume', ascending=False)
Temp_buy = tools.MixDataFrames(Temp_buy0)
if Temp_buy0.__contains__('price') and Temp_buy0['price'].empty == False:
Temp_buy = Temp_buy.sort_values(by='price', ascending=False)
if Temp_buy0.__contains__('volume') and Temp_buy0['volume'].empty == False:
Temp_buy = Temp_buy.sort_values(by='volume', ascending=False)
userInfo.buy_all_stock(Temp_buy)
#更新資訊--------------------------------------
if Temp_change <= 0 and len(userInfo.handle_stock) > 0:
Temp_change = mainParament.change_days
userInfo.Record_userInfo()
userInfo.Recod_tradeInfo()
Temp_result_pick = Temp_result_pick.append({'date':userInfo.now_day,
'選股數量':len(Temp_result)},ignore_index = True)
#加一天----------------------------
if userInfo.add_one_day() == False:
break
else:
Temp_change = Temp_change - 1
#最後總結算----------------------------
Temp_result_pick.set_index('date',inplace=True)
Temp_alldata = tools.MixDataFrames({'draw':userInfo.Temp_result_draw,'pick':Temp_result_pick},'date')
draw_backtest(userInfo.Temp_result_draw)
Temp_alldata = tools.MixDataFrames({'all':Temp_alldata,'userinfo':userInfo.Temp_result_All},'date')
userInfo.Temp_result_draw.to_csv('backtestdata.csv')
userInfo.Temp_trade_info.to_csv('backtesttrade.csv')
Temp_alldata.to_csv('backtestAll.csv')